How AI Learns New Tasks Using Old Data Labels
A method for helping AI models understand new topics by grouping similar labels from different datasets into a shared, broader category.
Original patent title: “Transfer learning techniques for disparate label sets”
A method for helping AI models understand new topics by grouping similar labels from different datasets into a shared, broader category. Granted to Microsoft Technology Licensing LLC in 2021 with 23 claims and 4 forward citations.
Key facts
Coverage
What does this patent actually cover?
This patent describes a way to teach an AI model about a new subject by using knowledge from a subject it already knows. It works by turning labels (like 'weather' or 'flight status') into mathematical vectors, which are lists of numbers that represent their meaning. The system then uses clustering algorithms to find common ground between these different labels, creating a 'coarse label' that acts as a bridge. For example, if the AI knows 'book flight' and 'reserve seat,' it can use this shared 'coarse' category to better understand a new, related task like 'check-in for flight' even if it has very little training data for that specific task.
The gap
What does this patent NOT cover?
- Does not cover systems that rely solely on manually defined rules rather than vector-based clustering.
- Does not cover models that perform transfer learning without first creating a coarse label set as an abstraction layer.
- Does not cover the specific hardware architecture, only the software-based method of label mapping.
These exclusions are unique to PatentBrief — derived from the actual claim language, not patent-office boilerplate.
What made this novel
Instead of trying to map labels directly from one domain to another, it creates an intermediate 'coarse' layer. This abstraction acts as a translator, allowing the model to see the semantic relationship between labels that might look completely different on the surface.
Schematic visualization of the patent's claim structure. Hand-drawn diagrams in progress for each landmark patent.
Where you've seen this
Real-world examples
Virtual assistants like Microsoft Cortana or Alexa
Customer service chatbots
Natural language understanding modules in search engines
Why it matters
The bigger picture
Training AI models from scratch is expensive and requires massive amounts of data. This technique allows companies to build smarter virtual assistants and chatbots by recycling knowledge across different domains, making AI more efficient and capable of handling new user requests without needing millions of new examples.
Filed
July 6, 2015
Granted
July 13, 2021
Market context
Who's building on this
Companies in this space
Microsoft is the primary assigneeassigneeThe entity that owns the patent — usually the inventor's employer or a company.Read more → and continues to integrate these techniques into their Azure AI and language processing services. Other major cloud providers like Google and Amazon actively research similar cross-domain transfer learning to improve their own voice and text-based AI models.
Market impact
This patent reflects the industry-wide shift toward data-efficient AI. By enabling transfer learning, it helps reduce the 'cold start' problem where new AI features struggle due to a lack of training data, effectively lowering the barrier for deploying sophisticated natural language tools across diverse business applications.
Claim 1 — Plain English
What this patent covers
This patent describes a way to teach an AI model about a new subject by using knowledge from a subject it already knows. It works by turning labels (like 'weather' or 'flight status') into mathematical vectors, which are lists of numbers that represent their meaning. The system then uses clustering algorithms to find common ground between these different labels, creating a 'coarse label' that acts as a bridge. For example, if the AI knows 'book flight' and 'reserve seat,' it can use this shared 'coarse' category to better understand a new, related task like 'check-in for flight' even if it has very little training data for that specific task.
The clever bit
Instead of trying to map labels directly from one domain to another, it creates an intermediate 'coarse' layer. This abstraction acts as a translator, allowing the model to see the semantic relationship between labels that might look completely different on the surface.
What it does not cover
- Does not cover systems that rely solely on manually defined rules rather than vector-based clustering.
- Does not cover models that perform transfer learning without first creating a coarse label set as an abstraction layer.
- Does not cover the specific hardware architecture, only the software-based method of label mapping.
Patent timeline
Application submitted to the patent office
Application published, typically 18 months after filing
Patent officially issued
PatentBrief Score
Impact Score
Strong
Citation count
14/40
Early citations
Claim breadth
15/20
Broad claimsclaimsThe numbered statements at the end of a patent that legally define what the inventor owns.Read more →
Recency
20/20
Granted within 5 years
Assignee scale
20/20
Major company or institution
PatentBrief Impact Score — based on citation count, claim breadth, recency, and assignee scale. Not a legal assessment.
Heuristic Value Estimate
What this patent might be worth
$78K – $250K
Midpoint $156K · 9.1 yr remaining · industry ×1.6
Heuristic only — blends forward/backward citation counts, claim scope, time remaining, litigation history, and CPC-derived industry baseline. Real valuations need a professional appraisal.
The original legal language
Original claims
23 claims as filed with the patent office.
Concepts involved
Citations
Patent lineage
Cite this patent
Kim, Y., & Sarikaya, R. (2021). How AI Learns New Tasks Using Old Data Labels (U.S. Patent No. 11,062,228). U.S. Patent and Trademark Office. https://patentbrief.org/patent/us/11062228/gpt-3-few-shot-learning
Auto-generated from the patent record. Double-check author order and the issue date against the official USPTO document before submitting.
Embed
Add this patent to your site
Drop this plain-English patent card into any blog post or article — free, no signup. It always links back to the full breakdown here.
<div data-patentlens-widget data-patent-number="US11062228"></div> <script src="https://patentbrief.org/embed.js" async></script>
Stay in the loop
Get a weekly digest of new patents.
One email per week. No spam. Unsubscribe anytime.
Keep exploring
Related patents you should know
US 4683195 · 1987
How to Make Billions of Copies of a DNA Segment
This patent describes the Polymerase Chain Reaction (PCR), a method to rapidly create many copies of a specific piece of DNA or RNA, enabling its detection and analysis.
Cetus Corp
US 8697359 · 2014
How to Edit Genes in Human Cells Using an Engineered CRISPR System
This patent describes an engineered CRISPR-Cas9 system for precisely cutting DNA in eukaryotic cells to change how genes work, opening the door for gene editing in complex organisms.
Massachusetts Institute of Technology
US 7657849 · 2010
How the iPhone's Slide-to-Unlock Gesture Works
Apple's 2010 patent describes unlocking a device by dragging a specific graphical image across the touchscreen along a predefined path, a gesture that became iconic with the original iPhone.
Apple Inc
US 4733665 · 1988
How Doctors Implant a Permanent Stent Using a Balloon
This patent describes the method for placing a permanent, expandable wire mesh tube inside a blood vessel or other body tube using a balloon-tipped catheter to widen it and keep it open.
Expandable Grafts Partnership
US 4965188 · 1990
How to Make Many Copies of a DNA Piece with Heat
This patent describes the Polymerase Chain Reaction (PCR) method, a technique to make millions of copies of a specific DNA segment using a heat-resistant enzyme and repeated temperature changes.
Cetus Corp
US 4235871 · 1980
How to Encapsulate Active Materials in Lipid Bubbles Efficiently
This patent describes a method for trapping biologically active substances inside tiny, multi-layered fat bubbles called liposomes, using a specific water-in-oil emulsion and gel-forming process to improve how much material gets captured.
Individual
More to explore
More in AI & Machine Learning
US 10452978 · 2019 · Google LLC
How AI Models Understand Language Using 'Attention'
US 6523026 · 2003 · Huntsman International LLC
How Computers Find Hidden Connections Between Different Fields of Knowledge
US 11615208 · 2023 · Capital One Services LLC
How Cloud Systems Automatically Create and Train AI Data Models
US 10402750 · 2019 · Facebook Inc
How Facebook Uses Deep Learning to Predict What You Might Like
New to patents?
Common Questions
Frequently Asked Questions
What does How AI Learns New Tasks Using Old Data Labels cover?
A method for helping AI models understand new topics by grouping similar labels from different datasets into a shared, broader category.
Who owns patent US 11062228?
Microsoft Technology Licensing LLC owns this patent, granted in 2021.
When does this patent expire?
This patent is expected to expire on July 13, 2041, when the invention enters the public domain.
What is patent US 11062228 cited by?
This patent has been cited by 4 later patents that build on its ideas.
What problem does this patent solve?
Training AI models from scratch is expensive and requires massive amounts of data. This technique allows companies to build smarter virtual assistants and chatbots by recycling knowledge across different domains, making AI more efficient and capable of handling new user requests without needing millions of new examples.
What does this patent NOT cover?
Does not cover systems that rely solely on manually defined rules rather than vector-based clustering.
Same assignee
More from Microsoft Technology Licensing LLC
Patent monitoring